Impact assessment of international peacebuilding projects has become a thorny issue at the nexus of policy research and academia. Ubiquitous demands and suggestions for sound methodologies notwithstanding (Menkhaus 2004; OECD 2012; Scharbatke-Church 2011; Scherrer 2012), scholars continue to have a hard time qualifying or quantifying what peacebuilding actually achieves. Classic social science methodology often seems inapt to measure the diffuse cause-and-effect relations of conflict settings, where observable changes are not easily attributable to the particular interventions at stake (Scherrer 2012: 14) and where the institutional timeframes for peacebuilding evaluations tend to be "considerably shorter than what is typically required to see tangible change" (Bush and Duggan 2014: 313). Finally, impact measurement continues to be primarily accountable to donors rather than to the populations affected by conflict, which makes it a crucial issue in the broader debates on peacebuilding and in whose name it is to be carried out (Barnett et al. 2007).
This paper investigates these concerns empirically and methodologically. Empirically, I focus on the case of Guinea, West Africa, where various international organizations and NGOs were financed by the UN Peacebuilding Fund to preserve the country's fragile peace during volatile periods of political turmoil since 2006. Having worked in and on Guinea since 2009, I reflect on the the failures and successes of some of these initiatives, as well as on the challenges of impact measurement. Taking the specific example of the NGO Search for Common Ground (SFCG), which has played an important role during the Ebola crisis in West Africa, I argue that current evaluation practices tend to find successes where there is little evidence, while ignoring those that actually exist on a broader structural level, which become visible only in the long run and/or in times of unexpected crisis. Relatedly, the paper problematizes the dominant formalism and lack of methodological reliability in impact measurement while neglecting longitudinal analyses and in-depth descriptions of actual processes between different stakeholders (Bächtold 2015; Scharbatke-Church 2011: 460). As an alternative, I suggest a more dynamic, process-oriented framework revolving around the notion of 'crystallization' (see Simondon 1958, 1989, 1995; Combes 2013). Such an approach has proven to constitute a fruitful basis for studying contemporary peace and conflict outcomes in fluid settings (Banegas 1993; Philipps 2016; Philipps and Kagoro 2016), and could inspire important new perspectives on international peacebuilding and how to measure its impacts.
